Measuring lymphocyte response to mitogens and antigens is a mainstay of screening for cellular immunodeficiency. Few reports analyze performance as a screening tool in diverse patient cohorts. We studied proliferation assays performed at Children's Hospital Boston from 1996 to 2003 using mitogens phytohemagglutinin (PHA), concanavalin A (CONA) and pokeweed mitogen, and antigens tetanus (TT) and diphtheria (DT) toxoids, and compared a subset of patients with T cell dysfunction with adult controls using receiver operating characteristic analysis. Results were correlated with clinical data. CONA was superior to PHA in identifying patients with immunodeficiency. TT was second best. Interpretation based on raw CPM, a stimulation index, or reference to simultaneous controls all performed equally. Combining data from multiple mitogens and/or antigens did not enhance performance. Proliferation testing is a useful component of screening for cellular immunodeficiency, but is not a sensitive predictor of cellular immune compromise or risk of opportunistic infection.